In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
Make sure that you've downloaded the required human and dog datasets:
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dogImages.
Download the human dataset. Unzip the folder and place it in the home diretcory, at location /lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("lfw/*/*/*.jpg"))
dog_files = np.array(glob("dogImages/*/*/*/*.jpg"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
Question 1: Use the code cell below to test the performance of the face_detector function.
human_files have a detected human face? dog_files have a detected human face? Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: (You can print out your results and/or write your percentages in this cell)
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
# for human_file in human_files_short:
human_as_human = 0
dog_as_human = 0
for h in tqdm(range(len(human_files_short)), ascii=True):
if face_detector(human_files_short[h]):
human_as_human += 1
human_pct = human_as_human / 100
for d in tqdm(range(len(dog_files_short)), ascii=True):
if face_detector(dog_files_short[d]):
dog_as_human += 1
dog_pct = dog_as_human / 100
print(f'Human detected as human: {human_pct * 100}%')
print(f'Dog detected as human: {dog_pct * 100}%')
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
Using CNN to detect human face.
The shape of images in the dataset is (3, 224, 224), while in dog dataset the shape varies.
import torch
import torch.nn as nn
import torch.nn.functional as F
# Define NN architecture to distinguish human and dog
class HumanFaceDetector(nn.Module):
def __init__(self):
super().__init__()
# convolutional layers
self.conv1 = nn.Conv2d(3, 16, 3, stride=1, padding=1)
self.conv2 = nn.Conv2d(16, 32, 3, stride=1, padding=1)
self.conv3 = nn.Conv2d(32, 64, 3, stride=1, padding=1)
self.conv4 = nn.Conv2d(64, 128, 3, stride=1, padding=1)
self.conv5 = nn.Conv2d(128, 256, 3, stride=1, padding=1)
# max pooling layer
self.pool = nn.MaxPool2d(2, 2)
# dropout layer
self.dropout = nn.Dropout(0.4)
# batch normalization layer
self.conv_bn1 = nn.BatchNorm2d(16)
self.conv_bn2 = nn.BatchNorm2d(32)
self.conv_bn3 = nn.BatchNorm2d(64)
self.conv_bn4 = nn.BatchNorm2d(128)
self.conv_bn5 = nn.BatchNorm2d(256)
# fully connected layer
self.fc1 = nn.Linear(7 * 7 * 256, 512)
self.fc2 = nn.Linear(512, 256)
self.fc3 = nn.Linear(256, 64)
self.fc4 = nn.Linear(64, 2)
def forward(self, x):
# add sequence of convolutional and max pooling layers
x = self.pool(F.relu(self.conv1(x)))
x = self.conv_bn1(x)
x = self.pool(F.relu(self.conv2(x)))
x = self.conv_bn2(x)
x = self.pool(F.relu(self.conv3(x)))
x = self.conv_bn3(x)
x = self.pool(F.relu(self.conv4(x)))
x = self.conv_bn4(x)
x = self.pool(F.relu(self.conv5(x)))
x = self.conv_bn5(x)
# flatten image input
x = x.view(-1, 256 * 7 * 7)
# add dropout layer
x = self.dropout(x)
x = F.relu(self.fc1(x))
x = self.dropout(x)
x = F.relu(self.fc2(x))
x = self.dropout(x)
x = F.relu(self.fc3(x))
x = self.fc4(x)
return x
# initialize the NN
hfdetector = HumanFaceDetector()
print(hfdetector)
if torch.cuda.is_available():
hfdetector = hfdetector.cuda()
# image preprocessing function
from PIL import Image
def shrink_large_img(img_path):
bad_file_list = []
with Image.open(img_path) as img:
if min(img.size) > 512:
try:
print(f'File {img_path} shrinked!')
img.thumbnail((275, 275))
img.save(img_path, 'JPEG')
except OSError as e:
print(f'Bad image: {img_path}')
bad_file_list.append(img_path)
for file in bad_file_list:
os.remove(file)
print(f'{file} is bad and removed!')
Create dataset for human and dog:
# create dataset for human and dog (Could this setting result in overfitting?)
# only run first time
import os
import shutil
from random import shuffle
def createFolder(directory):
try:
if not os.path.exists(directory):
os.makedirs(directory)
except OSError:
print ('Error: Creating directory. ' + directory)
# create image folders
createFolder('cnn_dataset/train/human')
createFolder('cnn_dataset/valid/human')
createFolder('cnn_dataset/test/human')
createFolder('cnn_dataset/train/dog')
createFolder('cnn_dataset/valid/dog')
createFolder('cnn_dataset/test/dog')
# copy images to above folders
# train : validate : test = 6 : 2 : 2
human_files, dog_files = list(human_files), list(dog_files)
shuffle(human_files)
shuffle(dog_files)
human_train = human_files[:int(len(human_files) * 0.6)]
human_valid = human_files[int(len(human_files) * 0.6): int(len(human_files) * 0.8)]
human_test = human_files[int(len(human_files) * 0.8):]
dog_train = dog_files[:int(len(dog_files) * 0.6)]
dog_valid = dog_files[int(len(dog_files) * 0.6): int(len(human_files) * 0.8)]
dog_test = dog_files[int(len(dog_files) * 0.8):]
# copy files
if not os.listdir('cnn_dataset/train/human'):
for dir in human_train:
shutil.copy(dir, 'cnn_dataset/train/human')
for dir in human_valid:
shutil.copy(dir, 'cnn_dataset/valid/human')
for dir in human_test:
shutil.copy(dir, 'cnn_dataset/test/human')
for dir in dog_train:
shutil.copy(dir, 'cnn_dataset/train/dog')
for dir in dog_valid:
shutil.copy(dir, 'cnn_dataset/valid/dog')
for dir in dog_test:
shutil.copy(dir, 'cnn_dataset/test/dog')
Data loader:
from torchvision import datasets, transforms
# Hyperparameters:
batch_size = 128
num_workers = 0
# data root
train_dir = 'cnn_dataset/train'
valid_dir = 'cnn_dataset/valid'
test_dir = 'cnn_dataset/test'
# preprocess images: shrink too big images:
human_files = np.array(glob('cnn_dataset/*/human/*.jpg'))
dog_files = np.array(glob('cnn_dataset/*/dog/*.jpg'))
for path in np.concatenate((human_files, dog_files)):
shrink_large_img(path)
# training data transforms
train_transforms = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
])
# validation and test data transforms
vt_transforms = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
])
# Pass transforms in here, then run the next cell to see how the transforms look
train_data = datasets.ImageFolder(train_dir, transform=train_transforms)
valid_data = datasets.ImageFolder(valid_dir, transform=vt_transforms)
test_data = datasets.ImageFolder(test_dir, transform=vt_transforms)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
Visualize a batch data:
dataiter = iter(train_loader)
images, labels = dataiter.next()
images = images.numpy() # convert images to numpy for display
classes = ['dog', 'human']
# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(25, 4))
for idx in np.arange(10):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
plt.imshow(np.transpose(images[idx], (1, 2, 0)))
ax.set_title(classes[labels[idx]])
Class to measure data loader time
# class to measure data loader time
class AverageMeter:
"""Computes and stores the average and current value"""
def __init__(self):
self.reset()
def reset(self):
self.val = 0
self.avg = 0
self.sum = 0
self.count = 0
def update(self, val, n=1):
self.val = val
self.sum += val * n
self.count += n
self.avg = self.sum / self.count
Train the model:
import time
import torch.optim as optim
# specify loss function (categorical cross-entropy)
criterion = torch.nn.CrossEntropyLoss()
# if model already trained, then just load it:
if os.path.exists('human_dog_distinctor.pt'):
print('Shall load pretrained model in next cell.')
else:
torch.cuda.empty_cache()
# specify optimizer (stochastic gradient descent) and learning rate = 0.001
optimizer = optim.Adam(hfdetector.parameters(), lr=0.00003)
start = time.time()
print(f'Training started at {time.ctime()}')
# number of epochs to train the model
n_epochs = 50
stop_criterion = 5
valid_loss_min = np.Inf
early_stop_count = 0
if torch.cuda.device_count() >= 2:
print("Let's use", torch.cuda.device_count(), "GPUs!")
hfdetector = torch.nn.DataParallel(hfdetector)
elif torch.cuda.is_available():
hfdetector = hfdetector.cuda()
# Time meter
batch_time = AverageMeter()
data_time = AverageMeter()
for epoch in range(1, n_epochs+1):
# keep track of training and validation loss
train_loss = 0.0
valid_loss = 0.0
# early stop mechanism:
if early_stop_count >= stop_criterion:
print(f'Validation loss stops decresing for {stop_criterion} epochs, early stop triggered.')
break
###################
# train the model #
###################
hfdetector.train()
e = time.time()
for data, target in train_loader:
# measure data loading time
data_time.update(time.time() - e)
# move tensors to GPU if CUDA is available
if torch.cuda.is_available():
data, target = data.cuda(), target.cuda()
# clear the gradients of all optimized variables
optimizer.zero_grad()
# forward pass: compute predicted outputs by passing inputs to the model
output = hfdetector(data)
# calculate the batch loss
loss = criterion(output, target)
# backward pass: compute gradient of the loss with respect to model parameters
loss.backward()
# perform a single optimization step (parameter update)
optimizer.step()
# update training loss
train_loss += loss.item() * data.size(0)
# measure elapsed time
batch_time.update(time.time() - e)
e = time.time()
######################
# validate the model #
######################
hfdetector.eval()
for data, target in valid_loader:
# move tensors to GPU if CUDA is available
if torch.cuda.is_available():
data, target = data.cuda(non_blocking=True), target.cuda(non_blocking=True)
# forward pass: compute predicted outputs by passing inputs to the model
output = hfdetector(data)
# calculate the batch loss
loss = criterion(output, target)
# update average validation loss
valid_loss += loss.item() * data.size(0)
# calculate average losses
train_loss = train_loss/len(train_loader.dataset)
valid_loss = valid_loss/len(valid_loader.dataset)
# print training/validation statistics
print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
epoch, train_loss, valid_loss))
# save model if validation loss has decreased
if valid_loss < valid_loss_min:
print('Validation loss decreased ({:.6f} --> {:.6f}). Saving model ...'.format(
valid_loss_min,
valid_loss))
torch.save(hfdetector.state_dict(), 'human_dog_distinctor.pt')
valid_loss_min = valid_loss
early_stop_count = 0
else:
early_stop_count += 1
# print time
print(f'Time {batch_time.val:.3f}s ({batch_time.avg:.3f}s) Data {data_time.val:.3f}s ({data_time.avg:.3f}s)')
end = time.time()
t = int(end - start)
print(f'Training ended at {time.ctime()}, total training time is {t//3600}hours {(t%3600)//60}minutes {(t%3600)%60} seconds.')
# load model with lowest validation loss
hfdetector.load_state_dict(torch.load('human_dog_distinctor.pt'))
# track test loss
test_loss = 0.0
class_correct = list(0. for i in range(2))
class_total = list(0. for i in range(2))
hfdetector.eval()
# iterate over test data
for data, target in test_loader:
# move tensors to GPU if CUDA is available
if torch.cuda.is_available():
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = hfdetector(data)
# calculate the batch loss
loss = criterion(output, target)
# print(f'loss -> {loss} output -> {F.softmax(output)} target -> {target}')
# update test loss
test_loss += loss.item() * data.size(0)
# convert output probabilities to predicted class
_, pred = torch.max(output, 1)
# compare predictions to true label
correct_tensor = pred.eq(target.data.view_as(pred))
correct = np.squeeze(correct_tensor.numpy()) if not torch.cuda.is_available() else np.squeeze(correct_tensor.cpu().numpy())
# calculate test accuracy for each object class
for i in range(batch_size):
if i >= target.data.shape[0]:
break
label = target.data[i]
class_correct[label] += correct[i].item()
class_total[label] += 1
# average test loss
test_loss = test_loss/len(test_loader.dataset)
print('Test Loss: {:.6f}\n'.format(test_loss))
for i in range(2):
if class_total[i] > 0:
print('Test Accuracy of %5s: %2d%% (%2d/%2d)' % (
classes[i], 100 * class_correct[i] / class_total[i],
np.sum(class_correct[i]), np.sum(class_total[i])))
else:
print('Test Accuracy of %5s: N/A (no training examples)' % (classes[i]))
print('\nTest Accuracy (Overall): %2d%% (%2d/%2d)' % (
100. * np.sum(class_correct) / np.sum(class_total),
np.sum(class_correct), np.sum(class_total)))
import matplotlib.pyplot as plt
%matplotlib inline
# helper function to un-normalize and display an image
def imshow(img):
if torch.cuda.is_available():
img = img.cpu()
img = img / 2 + 0.5 # unnormalize
plt.imshow(np.transpose(img, (1, 2, 0))) # convert from Tensor image
# obtain one batch of test images
dataiter = iter(test_loader)
images, labels = dataiter.next()
images.numpy()
# move model inputs to cuda, if GPU available
if torch.cuda.is_available():
images = images.cuda(non_blocking=True)
# get sample outputs
output = hfdetector(images)
# convert output probabilities to predicted class
_, preds_tensor = torch.max(output, 1)
preds = np.squeeze(preds_tensor.numpy()) if not torch.cuda.is_available() else np.squeeze(preds_tensor.cpu().numpy())
# plot the images in the batch, along with predicted and true labels
fig = plt.figure(figsize=(25, 4))
for idx in np.arange(20):
ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
imshow(images[idx])
ax.set_title("{} ({})".format(classes[preds[idx]], classes[labels[idx]]),
color=("green" if preds[idx]==labels[idx].item() else "red"))
### (Optional)
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
import torchvision.transforms.functional as TF
def cnn_face_detector(img_path):
'''
Detector to dectect human face using a trained deep learning model.
Return the index of prediction.
Params:
img_path: A path of an image
'''
classifier = HumanFaceDetector()
# because the state_dict is saved as a model that trained under parallel mode:
classifier = torch.nn.DataParallel(classifier)
classifier.load_state_dict(torch.load('human_dog_distinctor.pt'))
image = Image.open(img_path)
# Transforms:
image = TF.resize(image, 256)
center_crop = transforms.CenterCrop(224)
image = center_crop(image)
image = TF.to_tensor(image)
normalize = transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225))
image = normalize(image)
image = image.view(1, *tuple(image.shape))
# add additional batch dimension for singleton image:
if torch.cuda.is_available():
classifier = classifier.cuda()
image = image.cuda()
output = classifier(image)
_, pred = torch.max(output, 1)
return pred
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
## TODO: Test the performance of the face_detector algorithm
## on the images in human_files_short and dog_files_short.
# for human_file in human_files_short:
human_as_human = 0
dog_as_human = 0
for h in tqdm(range(len(human_files_short)), ascii=True):
if cnn_face_detector(human_files_short[h]):
human_as_human += 1
human_pct = human_as_human / 100
for d in tqdm(range(len(dog_files_short)), ascii=True):
if cnn_face_detector(dog_files_short[d]):
dog_as_human += 1
dog_pct = dog_as_human / 100
print(f'Human detected as human: {human_pct * 100}%')
print(f'Dog detected as human: {dog_pct * 100}%')
We can see that the accuracy has improved.
In this section, we use a pre-trained model to detect dogs in images.
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
import torchvision.transforms.functional as TF
def load_image(img_path, max_size=400, shape=None):
''' Load in and transform an image, making sure the image
is <= 400 pixels in the x-y dims.'''
image = Image.open(img_path).convert('RGB')
# large images will slow down processing
if max(image.size) > max_size:
size = max_size
else:
size = max(image.size)
if shape is not None:
size = shape
in_transform = transforms.Compose([
transforms.Resize(size),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))])
# discard the transparent, alpha channel (that's the :3) and add the batch dimension
image = in_transform(image)[:3,:,:].unsqueeze(0)
if torch.cuda.is_available():
image = image.cuda()
return image
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
## TODO: Complete the function.
## Load and pre-process an image from the given img_path
## Return the *index* of the predicted class for that image
image = Image.open(img_path)
image = TF.resize(image, 250)
center_crop = transforms.CenterCrop(224)
image = center_crop(image)
image = TF.to_tensor(image)
normalize = transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
image = normalize(image)
image = image.view(1, *tuple(image.shape))
if torch.cuda.is_available():
image = image.cuda()
res = VGG16(image)
_, index = torch.max(res, 1)
return index
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
## TODO: Complete the function.
index = VGG16_predict(img_path)
# inclusive index range: [151, 268]
return ((index >= 151) and (index <= 268))
Question 2: Use the code cell below to test the performance of your dog_detector function.
human_files_short have a detected dog? dog_files_short have a detected dog?Answer:
from tqdm import tqdm
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
# Test the performance of the dog_detector algorithm
# on the images in human_files_short and dog_files_short.
human_as_dog = 0
dog_as_dog = 0
for h in tqdm(range(len(human_files_short)), ascii=True):
if dog_detector(human_files_short[h]):
human_as_dog += 1
human_pct = human_as_dog / 100
for d in tqdm(range(len(dog_files_short)), ascii=True):
if dog_detector(dog_files_short[d]):
dog_as_dog += 1
dog_pct = dog_as_dog / 100
print(f'Human detected as dog: {human_pct * 100}%')
print(f'Dog detected as dog: {dog_pct * 100}%')
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
### (Optional)
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
![]() |
![]() |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
![]() |
![]() |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
![]() |
![]() |
![]() |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import os
import numpy as np
import time
import copy
from glob import glob
import torch
import torchvision
import matplotlib.pyplot as plt
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.optim import lr_scheduler
from torchvision import datasets, models
import torchvision.transforms as transforms
from torch.utils.data.sampler import SubsetRandomSampler
from PIL import Image
from torch.autograd import Variable
import random
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
plt.ion() # interactive mode
%matplotlib inline
### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
# Hyperparameters:
batch_size = 64
num_workers = 0
# data root
train_dir = 'dogImages/train'
valid_dir = 'dogImages/valid'
test_dir = 'dogImages/test'
# preprocess images: shrink too big images:
raw_files = np.array(glob('dogImages/*/*/*.jpg'))
for path in raw_files:
shrink_large_img(path)
# training data transforms
train_transforms = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
])
# validation and test data transforms
vt_transforms = transforms.Compose([transforms.Resize(256),
transforms.CenterCrop(224),
transforms.RandomHorizontalFlip(),
transforms.RandomRotation(10),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
])
# Pass transforms in here, then run the next cell to see how the transforms look
train_data = datasets.ImageFolder(train_dir, transform=train_transforms)
valid_data = datasets.ImageFolder(valid_dir, transform=vt_transforms)
test_data = datasets.ImageFolder(test_dir, transform=vt_transforms)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
loaders_scratch = {'train': train_loader, 'valid': valid_loader, 'test': test_loader}
class_names = train_data.classes
n_classes = len(class_names)
print(n_classes)
def imshow(inp):
inp = inp.numpy().transpose((1, 2, 0))
mean = np.array([0.485, 0.456, 0.406])
std = np.array([0.229, 0.224, 0.225])
inp = std * inp + mean
inp = np.clip(inp, 0, 1)
plt.imshow(inp)
# Get a batch of training data
images, classes = next(iter(loaders_scratch['train']))
fig = plt.figure(figsize=(100,100))
for idx in np.arange(batch_size):
ax = fig.add_subplot(8, batch_size//8, idx+1, xticks=[], yticks=[])
imshow(images[idx])
# ax.set_title(class_names[classes[idx]].split(".")[1])